Cotton is one of the most valuable and widely grown cash crops internationally. Its annual production worldwide is over 100 million bales valued at US$45 billion. Asia is the biggest cotton production area, with four out of five world top cotton producers located in this region. Cotton is not only the main supporter for the textile industry, but it also provides a huge and profitable market for manufacturers of chemicals for weed, disease and pest control. There are diverse opportunities for cotton molecular improvement, including improvement of yield and fiber quality and creation of new varieties that are resistant to herbicides, insects, nematodes and diseases (Steward, 1991).
Tissue Culture of Cotton: In 1935, Skovsted reported the first embryo culture of cotton. Beasley (1971) reported callus formation in cotton as an outgrowth from the micropylar end of fertilized ovules on Murashige & Skoog (MS) medium. Somatic embryogenesis was achieved from a suspension culture of G. klotzschianum (Price & Smith, 1979). In 1983, Davidonis & Hamilton first succeeded in efficient and repeatable regeneration of cotton (G. hirsutum L.) plants from callus after two-year cultivation. Cotton plants were since regenerated through somatic embryogenesis from different explants (Zhang & Feng, 1992; Zhang, 1994) including cotyledon (Davidonis et al., 1987; Davidonis & Hamilton, 1983; Finer, 1988; Firoozabady et al., 1987), hypocotyl (Cousins et al., 1991; Rangan & Zavala, 1984; Rangan & Rajasekaran, 1996; Trolinder & Goodin, 1988; Umbeck et al., 1987, 1989), stem (Altman et al., 1990; Bajaj et al., 1989; Chen, et al. 1987; Finer & Smith, 1984), shoot apex (Bajaj et al., 1985; Gould et al., 1991; Turaev & Shamina, 1986), immature embryo (Beasley, 1971; Stewart & Hsu, 1977, 1978), petiole (Finer & Smith, 1984; Gawel et al., 1986; Gawel & Robacker, 1990), leaf (Finer & Smith, 1984; Gawel & Robacker, 1986), root (Chen & Xia, 1991; Kuo et al., 1989), callus (Finer & McMullen, 1990; Trolinder et al., 1991) and protoplast (Chen et al., 1989).
Transformation of cotton: Agrobacterium-mediated cotton transformation was first reported a decade ago with hypocotyl and cotyledon as explants (Firoozabady et al., 1987; Umbeck et al., 1987). Several useful genes have been introduced into cotton via Agrobacterium-mediated transformation, including insect and herbicide resistance genes (Perlak et al., 1990; Trolinder et al., 1991; Chen et al., 1994). Explants (such as hypocotyl, cotyledon, callus generated from hypocotyl and cotyledon, as well as immature embryos) have been used for Agrobacterium-mediated transformation and particle bombardment (de Framond et al., 1983; Finer & McMullen, 1990; Firoozabady et al., 1987; Perlak et al., 1990; Rangan & Rajasekaran, 1996; Rajasekaran et al., 1996; Trolinder et al., 1991; Umbeck et al., 1987, 1989, 1992). In addition, meristematic tissue of excised embryonic axes has also been used for cotton transformation by particle bombardment (Chlan et al., 1995; John, 1996; John & Keller, 1996; McCabe & Martinell, 1993). Zhou et al. (1983) transformed cotton by injecting DNA into the axile placenta one day after self-pollination.
However, the transformation rates were generally low, ranging from 20 to 30% when hypocotyl were used as explant (Firoozabady et al., 1987; Cousins et al., 1991; Rajasekaran et al., 1996). A significantly higher transformation efficiency, up to 80%, was reported when cotyledon was used as explant and the ocs gene encoding octopine synthetase used as the reporter gene (Firoozabady et al., 1987). However, the validity of octopine as a marker for transformation is questionable because octopine has been found in several plant species certainly not transformed by infection with A. tumefaciens (Wendt-Gallitelli and Dobrigkeit, 1973). A more recent report indicated that the transformation efficiency of cotyledon was about 20 to 30% (Cousins et al., 1991). The transformation efficiency was even lower when particle bombardment method was used (Keller et al., 1997). A difference in the type of explants used for transformation could have a significant effect on the efficiency of transformation and regeneration. It has been reported, for example, that for reducing false positive transformants, cotyledon was a better explant than hypocotyledon (Firoozabady et al., 1987).
Cotton transformation also is highly dependent on genotype (Trolinder, 1985a, 1986; Trolinder & Goodin, 1987, 1988a, 1988b; Trolinder & Chen, 1989). Apart from a few cultivars which are regenerable and transformable, such as Gossypium hirsutum cv. Coker 312 and G. hirsutum Jin 7, most other important elite commercial cultivars, such as G. hirsutum cv. D&P 5415 and G. hirsutum cv. Zhongmian 12, are not regenerable and transformable by these methods. The absence of a high-efficiency plant regeneration method has been regarded as a major obstacle to the application of Agrobacterium-mediated transformation to cotton (Gawel et al., 1986; Firoozabady et al., 1987).